摘要
为探究最大似然分类和面向对象分类两种方法在土地覆盖利用检测中的有效性和适用性,对沈阳市区土地利用检测方法进行对比研究。利用沈阳市区1989年、2000年、2010年和2016年4个时相Landsat遥感影像数据,将遥感影像土地覆盖/利用分类分为水域、建设用地、耕地、其他植被、其他用地5种类型。通过对比两次实验过程和结果的总体精度与Kappa系数,比较两种检测方法在实际应用中的优劣。基于像元的最大似然法和面向对象分类法总体精度分别优于80%、90%,Kappa系数分别为大于0.8、大于或接近0.9。面向对象影像分类方法精度高、效果理想,更适合高分辨率遥感影像的分类处理。
To investigate the effectiveness and applicability of maximum likelihood classification and object-oriented classification in land cover use detection,a comparative study was conducted on land use detection methods in the urban area of Shenyang City.Based on the Landsat remote sensing images of the urban area of Shenyang City in 1989,2000,2010 and 2016,the land cover/use classification of remote sensing images was divided into five types:water area,construction land,cultivated land,other vegetation,and other land.By comparing the overall accuracy and Kappa coefficient of the two experimental processes and results,it evaluated the strengths and weaknesses of the two detection methods in practical applications.The overall accuracy of the pixel-based maximum likelihood method and object-oriented classification method exceeded 80%and 90%,respectively,with Kappa coefficients greater than 0.8 and greater than or close to 0.9,respectively.The object-oriented image classification method demonstrates high accuracy and ideal results,making it more suitable for the classification processing of high-resolution remote sensing mages.
作者
杨大勇
刘晨红
张其红
彭秀侠
李海宇
YANG Dayong;LIU Chenhong;ZHANG Qihong;PENG Xiuxia;LI Haiyu(School of Transportation and Geomatics Engineering,Shenyang Jianzhu University,Shenyang Liaoning 110168,China)
出处
《河北环境工程学院学报》
CAS
2024年第6期79-84,共6页
Journal of Hebei University of Environmental Engineering
基金
沈阳市社会科学课题(SYSK2023-01-175)。
关键词
沈阳市区
土地利用
检测方法
比较研究
面向对象
Shenyang urban use
land cover
detection method
comparative study
object oriented
